update

CHANGELOG.md

@@ -1,6 +1,11 @@
 Changelog
 =========
 
+Version 0.88.10
+--------------
+* SVM C val defaults to 1
+* fixed bug in agender_agender
+
 Version 0.88.9
 --------------
 * added performance_weighted ensemble

nkululeko/constants.py

@@ -1,2 +1,2 @@
-VERSION="0.88.9"
+VERSION="0.88.10"
 SAMPLING_RATE = 16000

nkululeko/data/dataset.py

@@ -3,22 +3,21 @@
 import os
 import os.path
 from random import sample
+
 import numpy as np
 import pandas as pd
 
 import audformat
-from audformat.utils import duration
 
-import nkululeko.filter_data as filter
-import nkululeko.glob_conf as glob_conf
 from nkululeko.filter_data import DataFilter
+import nkululeko.glob_conf as glob_conf
 from nkululeko.plots import Plots
 from nkululeko.reporting.report_item import ReportItem
 from nkululeko.utils.util import Util
 
 
 class Dataset:
-    """Class to represent datasets"""
+    """Class to represent datasets."""
 
     name = ""  # An identifier for the dataset
     config = None  # The configuration

nkululeko/modelrunner.py

@@ -132,15 +132,15 @@ def _select_model(self, model_type):
         self._check_balancing()
 
         if model_type == "svm":
-            from nkululeko.models.model_svm import SVM_model
+            from nkululeko.models.model_svm import SVMModel
 
-            self.model = SVM_model(
+            self.model = SVMModel(
                 self.df_train, self.df_test, self.feats_train, self.feats_test
             )
         elif model_type == "svr":
-            from nkululeko.models.model_svr import SVR_model
+            from nkululeko.models.model_svr import SVRModel
 
-            self.model = SVR_model(
+            self.model = SVRModel(
                 self.df_train, self.df_test, self.feats_train, self.feats_test
             )
         elif model_type == "xgb":

nkululeko/models/model_svm.py

@@ -1,19 +1,19 @@
 # model_svm.py
 
-import random
 from sklearn import svm
+
 from nkululeko.models.model import Model
 
 
-class SVM_model(Model):
-    """An SVM model"""
+class SVMModel(Model):
+    """An SVM model."""
 
     is_classifier = True
 
     def __init__(self, df_train, df_test, feats_train, feats_test):
         super().__init__(df_train, df_test, feats_train, feats_test)
         self.name = "svm"
-        c = float(self.util.config_val("MODEL", "C_val", "0.001"))
+        c = float(self.util.config_val("MODEL", "C_val", "1"))
         if eval(self.util.config_val("MODEL", "class_weight", "False")):
             class_weight = "balanced"
         else:
@@ -25,7 +25,7 @@ def __init__(self, df_train, df_test, feats_train, feats_test):
             gamma="scale",
             probability=True,
             class_weight=class_weight,
-            random_state=42,    # for consistent result
+            random_state=42,  # for consistent result
         )  # set up the classifier
 
     def set_c(self, c):

nkululeko/models/model_svr.py

@@ -1,11 +1,12 @@
 # svrmodel.py
 
 from sklearn import svm
+
 from nkululeko.models.model import Model
 
 
-class SVR_model(Model):
-    """An SVR model"""
+class SVRModel(Model):
+    """An SVR model."""
 
     is_classifier = False
 

tests/exp_emodb_os_svm.ini

@@ -4,7 +4,8 @@ name = exp_emodb_classifiers
 save = True
 [DATA]
 databases = ['emodb']
-emodb = ./data/emodb/emodb
+#emodb = ./data/emodb/emodb
+emodb = /home/audeering.local/fburkhardt/Downloads/emodb
 emodb.split_strategy = specified
 emodb.test_tables = ['emotion.categories.test.gold_standard']
 emodb.train_tables = ['emotion.categories.train.gold_standard']